This invention relates to a surface position detecting method for detecting level (height) or tilt of a surface of an object such as a wafer, for example, having a region with a pattern structure. More particularly, the invention is concerned with a surface position detecting method usable in a slit scan type exposure apparatus, for example, for continuously detecting position or tilt of a surface of a wafer, for example, with respect to an optical axis direction of a projection optical system of the exposure apparatus.
The size of recent memory chips is increasing because of a difference between the trend of cell size or resolvable line width of an exposure apparatus and the enlargement trend of memory capacity. For a first generation of 256 M, for example, it is reported that the size is about 14xc3x9725 mm.
For this chip size and with an exposure region of a diameter 31 mm of a reduction projection exposure apparatus (stepper) which is currently used as an exposure apparatus for a critical layer, only one chip can be exposed per one exposure operation and the throughput is low. Thus, an exposure apparatus which enables a larger exposure area is required. For such a large field size exposure apparatus, there may be a semiconductor device exposure apparatus for a rough layer for which a high throughput is required or a reflection projection type exposure apparatus which is an exposure apparatus for a large picture field size liquid crystal display device such as a monitor. Such an apparatus is a slit scan exposure apparatus of a mask-to-wafer relative scan type wherein a mask is rectilinearly scanned with an arcuate slit-like illumination light and, by use of a concentric reflection mirror optical system, a wafer is exposed as a whole.
For focus adjustment of a mask image in such apparatus, level measurement and corrective drive for autofocusing or auto-leveling are performed continuously during the scan exposure operation so that the surface, to be exposed, of a photosensitive substrate (wafer or glass plate which is coated with a photoresist, for example) is brought into registration with a best imaging plane of a projection optical system.
A height and surface position detecting mechanism in such apparatus may use a method in which an oblique projection optical system for projecting light onto the surface of a wafer obliquely from above and reflection light from the photosensitive substrate is detected as a positional deviation upon a sensor, or a method in which a gap sensor such as an air microsensor or an electrostatic capacity sensor is used. From plural measured values during the scan, the amount of corrective drive for height and tilt as the measurement position passes the exposure slit region is calculated.
If only a projection system of a slit scan type exposure apparatus currently used is modified to provide a resolving power that meets 256 M or larger, the following problems arise.
As the numerical aperture (N.A.) of the reduction optical system is enlarged to meet miniaturization of a circuit pattern, the tolerable depth of focus during a circuit pattern transfer process is narrowed. Exposure apparatuses currently used for a rough layer process have a tolerable depth not less than 5 microns. Therefore, a measurement error included in measured values in continuous measurement during the scan exposure process or the effect of a surface step within a chip can be disregarded. However, for 245 M, the tolerable depth will be not greater than 1 micron. Thus, such measurement error or the effect of a surface step could not be disregarded. More specifically, when the focus (height and tilt) of a wafer surface is measured and focus correction is performed to keep the wafer surface within the tolerable depth, since the wafer surface contains surface irregularities, offset correction is necessary to bring the chip or shot as a whole in registration with the image plane. In such a case, unless the focus measurement point at each shot is the same as those during the offset measurement, accurate offset correction is not assured. It may be assured in a stepper in which measurement is performed as the wafer is stopped at each shot. However, in a scan system, it is not assured. When an accumulation type sensor is used and if the cycle of accumulation start is a free running type, there is an inaccuracy in position corresponding to the accumulation time, that is, a deviation between a focus measurement point and an offset measurement point. Thus, offset correction is inaccurate.
In a surface position detecting mechanism which uses an oblique light projection optical system, the reflectivity of a surface to be detected may change. A received light signal may be too strong and a light receiving system may be saturated. Alternatively, it may be too small and the signal-to-noise ratio (S/N ratio) may be degraded. It causes low surface position detection precision. Thus, during a scan exposure operation, it is necessary to adjust the gain (or light) of light receiving means or an emitted light quantity from light projecting means, in accordance with the reflectivity of the surface to be detected.
During continuous surface position measurement to a surface with different reflectivities during the scan, if the reflectivity of the surface is measured and the light is adjusted, the measurement time per one measurement operation may change and synchronism with respect to the stage position is broken. Thus, accurate offset correction is not attainable. Particularly, for an electric charge accumulation type sensor, an idle reading operation due to the light adjustment operation leads to a bottle neck of a response speed during the scan operation, and the throughput is lowered or there arise insufficiency of correction points. Thus, the surface position detection precision is degraded.
If a rectangular exposure area is sequentially transferred onto a wafer of a circular shape, as shown in FIGS. 1-3, at a peripheral portion of the wafer, a portion of the exposure area is out of the wafer region. In a wafer peripheral region, a multi-chip structure is adopted in which there are plural chips within the exposure area, this being to enable chip production even if a portion of the exposure area is missed (see FIG. 2). Thus, for such an exposure area in which a portion thereof is missing (non-rectangular portion), an ordinary exposure procedure has to be performed. In a structure in which one chip is present in an exposure region, such as a case of a CPU, in the process of wafer surface clamping such as in ion injection step or an RIE step, a photoresist remaining in a peripheral portion of the wafer may be separated and it may move to the pattern portion of the chip region, causing a pattern defect or dimension defect. It leads to degradation of the chip yield. It is, therefore, necessary to perform exposure to a non-rectangular portion to thereby remove any excessive resist thereon. Conventionally, during wafer exposure, the same exposure procedure is made to a non-rectangular portion at the peripheral portion of the wafer, as done to the rectangular portions inside the wafer.
In a slit scan type exposure apparatus, for exposure of a peripheral portion, focus is measured in real time and the wafer surface is corrected toward the lens image plane. Since, however, the exposure region (shot) is not rectangular as the wafer central portion, the focus measurement may end in failure and the exposure operation may be interrupted. Even if the error is detected and processed by use of software and the exposure process is continued, there remain an increase of processing time and synchronism delay with respect to a correction system. Thus, correction precision is not good.
It is an object of the present invention to provide a surface position detecting method by which the position of a surface to be detected can be detected precisely without being affected by a surface irregularity of the surface to be detected. Particularly, it may be applied to a slit scan exposure apparatus to ensure the synchronism with respect to position, in offset control for a focus measured value, and to assure high precision offset correction for a focus measured value and a high resolution pattern transfer operation.
It is another object of the present invention to assure improvement of surface position detection precision, by reducing or making constant the time for light adjustment during the scan measurement to thereby make constant the surface position measurement time during the scan. This enables the surface position detection in synchronism with the scan and accurate offset correction. Particularly, when it is applied to a slit scan exposure apparatus, the reflectivity of the region to be scan measured may be measured beforehand and the measurement cycle may be controlled constant, such that response delay with respect to synchronism with a servo system may be prevented.
It is a further object of the present invention to provide a surface position detecting system and method by which the point of surface position detection may be determined beforehand from the shape of the region to be processed, such that the position of that surface can be detected precisely without being affected by the shape of the region of the object to be detected. Particularly, it may be applied to high precision detection of focus, that is, the wafer surface position, in a slit scan type exposure system.
In accordance with an aspect of the present invention, there is provided a surface position detecting method wherein an object having a region with a pattern structure is relatively scanned relative to surface position detecting means and wherein surface position at plural detection points in the region is detected by use of the surface position detecting means, said method comprising the steps of: detecting an error in the detection by the surface position detecting means, with respect to each of the detection points, which error may result from a difference in pattern structure among the detection points; detecting, in synchronism with relative position of the object and the surface position detecting means, the surface position at each of the detection points by use of the surface position detecting means; and correcting the detected surface position on the basis of the error.
Said surface position detecting means may include light projecting means for projecting light obliquely onto the object and an accumulation type sensor for receiving reflection light from the object, and wherein said surface position detecting step includes detecting the surface position on the basis of the state of the reflection light.
Said surface position detecting step may include resetting the accumulation start timing of the sensor when the object and the surface position detecting means are placed in a predetermined relative position.
Said surface position detecting step may include driving the sensor and a scan control system for relatively scanning the object and the surface position detecting system, at the same clock.
When the surface position detection is to be done to plural objects having the same pattern structure, the error detection may be made with respect to a first one or ones of the objects.
In accordance with another aspect of the present invention, there is provided a surface position detecting system having surface position detecting means including light projecting means for projecting light obliquely onto a detection point and light receiving means for receiving reflection light from the detection point, wherein, while an object having a region with a pattern structure is relatively scanned relative to said surface position detecting means, a surface position at plural points within the region is detected, said surface position detecting system comprising: optimum value detecting means for projecting, before the surface position detection, light from said light projecting means to each of the plural detection points and for receiving, with said light receiving means, reflection light from the point, said optimum value detecting means detecting and memorizing an optimum value of a gain of said light receiving means or a drive current for said light projecting means with respect to each detection point, on the basis of a light reception signal of said light receiving means; and setting means for setting an optimum value for a drive current for said light projecting means or a gain of said light receiving means with respect to each of the detection points, for the surface position detection.
Said optimum value detecting means may include calculating means for receiving reflection light from each detection point while holding the drive current for the light projecting means and the gain of said light receiving means fixed in the state having been set before start of the detection, and for calculating an optimum value on the basis of the drive current, the gain and a light reception signal.
Said optimum value detecting means may detect an optimum value while scanning the drive current for the light projecting means and the gain of the light receiving means with respect to each of the detection points.
The object may have plural regions with the same pattern structure and wherein said optimum value detecting means may detect an optimum value on the basis of one or some of the plural regions.
Said optimum value detecting means may detect an optimum value while relatively scanning the object and said surface position detecting means relative to each other.
Said system may further comprise discriminating means for discriminating whether a detected optimum value is within a tolerable range, wherein the optimum value detection may be repeated when the result of the discrimination is negative.
Said light receiving means may include a one-dimensional CCD sensor and wherein said setting means may set an optimum value for each detection point and then reset the CCD sensor and start the surface position detection at that detection point.
In accordance with a further aspect of the present invention, there is provided a surface position detecting method wherein surface position detecting means having light projecting means for projecting light obliquely onto a detection point and light receiving means for receiving reflection light from the detection point are used and wherein, while relatively scanning an object having a region with a pattern structure relative to the surface position detecting means, a surface position with respect to plural detection points within the region is detected, said method comprising the steps of: projecting, before the surface position detection, light to each of the plural detection points and receiving, with the light receiving means, reflection light therefrom, and detecting and memorizing an optimum value of a drive current for the light projecting means or a gain of the light receiving means with respect to each detection point, on the basis of a light reception signal of the light receiving means; and performing the surface position detection while setting the drive current for the light projecting means or the gain of the light receiving means at the memorized optimum value, with respect to each of the detection points.
In accordance with a yet further aspect of the present invention, there is provided a surface position detecting system for detecting a surface position of a substrate, said system comprising: a plurality of sensors for measuring surface position at plural locations on the substrate, while relatively scanning the substrate; discriminating means for discriminating and memorizing, beforehand, effectiveness/ineffectiveness of each sensor at each measurement point during scanning measurement, on the basis of information related to the substrate processing; and calculating means for selecting, during scan measurement, one or those of the sensors of effective measured value on the basis of the discrimination information and for calculating the surface information of the substrate on the basis of a measurement output of the one or those sensors.
In accordance with a still further aspect of the present invention, there is provided a surface position detecting method wherein a plurality of sensors for measuring surface position of a substrate at plural locations while relatively scanning the substrate are used to detect a surface position corresponding to the measurement locations on the substrate, said method comprising the steps of: discriminating, beforehand, effectiveness/ineffectiveness of a measured value at each measurement location by a corresponding sensor during the scan measurement, on the basis of information related to the substrate processing; switching, during the scan measurement, sensors to be used for the measurement, dynamically, on the basis of the information of discrimination; and calculating the surface information of the substrate on the basis of a measurement output of the switched sensor or sensors.
In accordance with a yet further aspect of the present invention, there is provided an exposure method wherein a pattern of an original and a slit are projected onto a substrate through a projection optical system, wherein the original and the substrate are relatively scanned relative to the projection optical system in a direction perpendicular to a lengthwise direction of the slit whereby the pattern of the.original is transferred onto the substrate, and wherein a plurality of sensors for measuring surface position of the substrate at plural locations are used to detect surface position corresponding to the measurement locations on the substrate, said method comprising the steps of: discriminating, beforehand, effectiveness/ineffectiveness of a measured value of each of the measurement points of the sensors during the scan measurement, on the basis of information related to the substrate processing; switching, during the scan measurement, sensors to be used for the measurement, on the basis of the information of discrimination; calculating the surface information of the substrate on the basis of a measured output of the switched sensor or sensors; and determining, on the basis of the information of the discrimination, the order of shot processing so that, with regard to a shot with respect to which the number of effective sensors changes along the scan direction within that shot, the scan is done in a direction from a larger effective sensor number side to a smaller effective sensor number side.
These and other objects, features and advantages of the present invention will become more apparent upon a consideration of the following description of the preferred embodiments of the present invention taken in conjunction with the accompanying drawings.